There is no universal number of high-probability requests that should come first. Some published studies used three, but that procedural detail does not establish the right count for every person, task, or setting. A team should choose the shortest useful sequence from individualized data, define timing and consequences, monitor burden and withdrawal, compare outcomes with an appropriate baseline, and shorten, change, or stop the sequence when evidence warrants.
Choose how many high-probability requests fit
Start with the decision, not a memorized number. Consider response effort, total time, repetition, current health, preference, transition difficulty, partner skill, and the lower-probability request. A three-action sequence lasting ten seconds differs from three multi-step tasks.
Read study counts as study methods
The medical study placed three screened requests before each examination request. The variant-sequence study evaluated repeated versus varied pools in a different population and task. These procedures offer testable examples without creating a standard count.
Compare benefit with burden
Track the lower-probability response, time to reach it, total requests, repeated prompts, client report, withdrawal, distress, and disruption of the routine. More completed warm-ups can coexist with a slower, less comfortable, or less useful experience.
Use professional review
The BACB outline covers antecedent interventions, measurement, evaluation, contextual fit, and unwanted-effect mitigation. The Ethics Code addresses client involvement, assent when applicable, risk, and data-based evaluation. ASHA supports continued AAC access at every sequence length.
A practical example
Eli's team compares one and three warm-up requests across eight eligible opportunities each. The target response occurs in 5/8 after one and 5/8 after three. The three-request version takes longer and Eli selects one when offered a choice. The team adopts one and continues review.
Questions families can use
Why this count? How long does it take? Does the person choose or tolerate the sequence? Is the later response stronger? Which measure would justify adding or removing a request?
Sources
- Behavior Analyst Certification Board, BCBA Test Content Outline, 6th edition
- Shillingsburg and colleagues, Increasing Compliance With Medical Examination Requests Directed to Children With Autism
- Davis and Reichle, Variant and Invariant High-Probability Requests
- Mace and colleagues, Effects of Reinforcer Quality on Behavioral Momentum
- Behavior Analyst Certification Board, Ethics Code for Behavior Analysts
- American Speech-Language-Hearing Association, Augmentative and Alternative Communication
Finni resources
